Backlog plan
Break down a goal into actionable tasks with dependencies, priorities, and optional specs. Use when the user wants to plan work, decompose a feature, or create a task breakdown.
git clone https://github.com/backloghq/backlog
T=$(mktemp -d) && git clone --depth=1 https://github.com/backloghq/backlog "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/plan" ~/.claude/skills/backloghq-backlog-plan && rm -rf "$T"
skills/plan/SKILL.mdPlan Work
Break down the given goal into actionable tasks. The goal is: "$ARGUMENTS"
Process
-
Understand the goal — analyze what "$ARGUMENTS" requires. Read relevant code, configs, and docs in the codebase to understand the current state.
-
Decompose into tasks — create a logical breakdown of work items. Each task should be:
- Small enough to complete in one focused session
- Clear about what "done" looks like
- Independent where possible, with explicit dependencies where not
-
Create tasks — for each task, use
with:task_add- A clear, actionable description (start with a verb)
set to a logical groupingproject
based on urgency and importance (H for blockers, M for core work, L for nice-to-have)priority
for categorization (e.g.,tags
,frontend
,backend
,testing
)docs
to express ordering constraints (use UUIDs from previously created tasks)depends
if the task shouldn't start until a certain datescheduled
-
Write specs for complex tasks — for any task that needs detailed requirements or context, use
to attach a markdown document explaining:task_doc_write- What needs to be built
- Key technical decisions
- Acceptance criteria
- Relevant code locations
-
Present the plan — show the created tasks in dependency order with their IDs, descriptions, and priorities. Highlight the first task(s) that can be started immediately (not blocked).